Date:         Tue, 1 May 2001 10:53:56 -0700
Reply-To:     svpetrus@NEWWESTENERGY.COM
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Petruso Steven V <svpetrus@NEWWESTENERGY.COM>
Subject:      Re: Hot fixes, quality control, etc.
Comments: To: RHOADSM1@WESTAT.COM
Content-Type: text/plain; charset="iso-8859-1"

A little birdie just sent me an email ID to vent these issues to, here is
the text from the note:


"If you send email to suggest@sas.com, your message will be
available for all SAS employees to read."

         - Little Birdie

I'm going to send a note.

-----Original Message-----
From: Mike Rhoads [mailto:RHOADSM1@WESTAT.COM]
Sent: Tuesday, May 01, 2001 8:31 AM
To: SAS-L@LISTSERV.UGA.EDU
Subject: Hot fixes, quality control, etc.


SPECIFIC QUESTION:

There have been a large number of "hot fixes" to version 8.1 of SAS.  For
many of them, the associated SAS Note(s) indicate that the underlying
problem is either not fixed yet, or is perhaps fixed in version 9 or in
"TSLEVEL 2M0".  So, if someone is (1) running 8.1 with all fixes applied,
and (2) receives and installs 8.2, does this wind up "breaking" these hot
fixes?  Will there be parallel hot fixes for 8.2 that we should await?

RELATED NOTES/GRIPES:

Anybody else out there besides ourselves getting confused, confounded and
concerned about the blizzard of problems and hot fixes with recent SAS
software, particularly 8.1?

First, is it just me, or has the number of serious problems seemed to
increase with recent versions?  We certainly expected V7 to have a lot of
things floating around, since it was the first release with much of the code
rewritten to handle long var names and lengths, ODS, etc.  However, we had
hoped that things would settle down greatly with 8.0 (which we considered
more of a "7.1"), 8.1, and subsequent releases.  So far, this has not seemed
to be the case.  8.1 for instance, even just in Base SAS, seems to have
quite a few "serious" problems, which to me is some basic functionality that
doesn't work, or some reasonably plausible code that produces incorrect
output.

The SAS folks have certainly been responsive to dealing with these (many of
them, at least) once they are found: if I'm reading the table on the SAS web
site (http://ftp.sas.com/techsup/download/hotfix/v81fp.html) correctly,
there have been over 30 hot fixes just to Base SAS, judging from the
numbering scheme.

But, of course, dealing with all of these fixes, especially in a
multi-platform environment, presents its own set of problems.  Even ignoring
the newer products, this seems like much more of an effort than it was
several years ago.

Any thoughts or advice would be appreciated ...

Mike Rhoads
Westat
RhoadsM1@Westat.com